backtop


Print 72 comment(s) - last by Starcub.. on Sep 25 at 10:20 AM

Not many motherboards support new standard yet

Seagate is launching today the latest in the  Barracuda family of 7200 RPM hard disk drives. The Barracuda XT is the first drive to market that supports SATA interface speeds of up to 6Gbps. 

The 2TB monster features a large 64 MB cache, which is the largest seen on a regular HDD. However, several SSDs already use 128MB caches, and at least one controller design in the works is capable of accessing up to 256MB of cache. It is these large and fast caches that are driving SATA standards forward.
 
The latest version of Seagate's SeaTools software allows for short-stroking, in which data is stored only on the outer tracks of the drive, allowing greater access speed at a reduction in capacity. The company claims that a short-stroked Barracuda XT using 1TB of storage will be able to compete with a 10k RPM Velociraptor drive from competitor Western Digital.

The company is targeting high performance and gaming PCs, low cost servers for SMBs, and external storage applications using eSATA for the new drive. Seagate expects almost 20% of all HDDs sold in 2010 will have a capacity of  1TB or greater.

The new drive (model ST32000641AS) comes with a 5-year warranty at a MSRP of $299. It should be available at retail by the end of this week.

Despite all the enthusiasm from Seagate, it will be SSDs that see the greatest performance jump with the move to the next generation for the SATA interface. Several SSDs are already hitting the limits of SATA II when reading from their cache.

Adoption of the new SATA standard is currently slow, as the ASUS P7P55D is the only motherboard that is natively capable of support 6Gbps. Older motherboards are capable of such speeds only through the use of a PCIe adapter card.

The problem is that motherboard manufacturers are waiting for a new I/O Controller Hub (ICH) from Intel. Commonly known as a southbridge, the new ICH is expected to support new technologies such as SuperSpeed USB 3.0 and SATA 6Gbps. AMD is also working on a new southbridge to support these technologies.



Comments     Threshold


This article is over a month old, voting and posting comments is disabled

proofreading
By hessenpepper on 9/21/09, Rating: 0
RE: proofreading
By jonup on 9/21/09, Rating: -1
RE: proofreading
By TomZ on 9/21/2009 1:37:24 PM , Rating: 4
Your brain might ignore them, but for many of us, simple spelling and grammar errors undermine the credibility of the author. After all, if the author can't spell or proofread, how can he be trusted to get the facts right or to know what he's talking about on the subject?

One of the problems with the current Internet IMO is that there don't seem to be very professional standards set on the part of content publishers as much. Yet these publishers are still companies getting paid to do this work, and the articles are being read by large numbers of readers. So why the slack?

I'm an engineer/manager, and if I turned in poor work to my customers and other stakeholders, I'd certainly hear about it.


RE: proofreading
By GaryJohnson on 9/21/2009 2:13:57 PM , Rating: 5
Someone with good spelling and grammatical accuracy can be just as factually accurate or inaccurate as someone with bad spelling and grammatical accuracy.

I agree it's unprofessional, but it doesn't have any bearing on the credibility of the author.


RE: proofreading
By slyadams on 9/21/2009 6:48:24 PM , Rating: 2
I can't agree. Surely the desire of a person to write an accurate and well researched aritcle would also stretch to eliminating simple spelling/grammatical errors?

If you were due to have heart surgery and the surgeon gave you some letters or information he had written and they were strewn with errors would it not make you a little concerned about his attention to detail?


RE: proofreading
By blwest on 9/21/2009 9:19:19 PM , Rating: 1
You forget one thing. These guys don't write the articles, they paraphrase, poorly. In addition, they don't research the articles, they simply put all the tech articles Anandtech readers might like in one place.

Many of these people have a personal agenda against certain hardware/software vendors. In addition, they most likely use IE because Firefox underlines spelling mistakes. There is no way any of these articles went into a legitimate word processor.

They forget that the average Anandtech reader is highly educated and will pick up on BS quickly. If you write like a 6th grader, I'll trust you like I would a 6th grader. =)


RE: proofreading
By gstrickler on 9/21/2009 8:02:35 PM , Rating: 4
quote:
Someone with good spelling and grammatical accuracy can be just as factually accurate or inaccurate as someone with bad spelling and grammatical accuracy.
They can be, yes. However, such errors call into question the quality of the work and the attitude or "work ethic" of the author. It suggests that the author may be "sloppy", "lazy", or "careless". If a "reporter" doesn't bother to use a spelling and grammar checker before publishing an article, how much should I trust his ability to accurately find and report the facts for the article? Still, it doesn't necessarily mean the information is incorrect.

Statistically, there is a strong correlation between the number of significant grammar errors and the accuracy of the information presented. That doesn't mean that one causes the other, only that they're likely to be found together. When two things are commonly found together, finding one suggests the likely presence of the other.

When I read a resume for a job applicant and there are spelling errors or obvious grammatical errors, I question the quality or work I would get from someone who is "careless" about their resume (which should present the applicant in the best possible way, as long as it's accurate). What quality of work should I expect from someone who didn't bother to use a spelling and grammar checker on their resume?

Spelling errors (where the "word" isn't even a word in the dictionary) are inexcusable in most instances simply because nearly everything is created on a computer now and nearly every "editor" routine includes a very good spell check capability. Typographical errors where the incorrect word is used won't be detected by a spell check capability, but should be detected by a grammar check capability.

I'm more forgiving of grammar errors, and particularly of the types of grammatical errors that are common due to a widespread misunderstanding of grammar rules. Others are personal preference, for instance, I could have said "grammatical rules" (adjective + noun), but I chose "grammar rules" (phrase used as a noun).

Now, technically inclined and creative people will commonly not "worry" about "details" like spelling or grammar, after all there are editors and proof-readers (and now software) that can worry about and correct those. That's not a problem, as long as you have an editor, proof-reader, and/or software that verifies the work before it's published. Unfortunately, those steps are either frequently being skipped, or the are of insufficient quality to correct the mistakes, so the mistakes end up getting published.


RE: proofreading
By afkrotch on 9/22/2009 2:08:40 AM , Rating: 3
It's called attention to detail. If you can't even pay attention to the details in your own article, who's to say you were paying attention in the information gathering process.


RE: proofreading
By Starcub on 9/25/2009 10:20:46 AM , Rating: 2
quote:
Someone with good spelling and grammatical accuracy can be just as factually accurate or inaccurate as someone with bad spelling and grammatical accuracy.

Aside from the fact that you've got the order wrong, I understand what you were trying to say. Personally, I don't mind spelling errors so much, but gramatical errors can be extremely frustrating: the use of an innapropriate term, or something as simple as a misplaced comma, can completely change the meaning of a sentence. If I have to re-read a sentence multiple times and attempt to guess what the author's actual message was from context, then I tire of reading the article, and even though the person might have had something important to say, I will ignore them.

I almost always make mistakes the first time through anything I compose, so I go back and re-read my posts to ensure that I'm able to understand what it is I was trying to say.


RE: proofreading
By luhar49 on 9/22/2009 2:04:55 PM , Rating: 2
Cut the author some slack. He is probably a non-native English speaker. Chinese, going by his name. Although that isn't an excuse when you post on an English website. Still, pointing the error is enough. Dont have to write an article of your own about bad english. Your post just sounds like a frustrated individual venting.


RE: proofreading
By plowak on 9/23/2009 2:54:18 PM , Rating: 2
Students, give me your attention. As an exercise in communication skills, I would like you to underline all of the non sequiturs in the above paragraph. You have five minutes...go!


RE: proofreading
By MrPoletski on 9/24/2009 10:02:12 AM , Rating: 2
Like I've said before.

You see a grammatical error, or something spelt incorrectly then:

Quote the section of text;
Highlight your percieved error;
Provide a suggested correction;
End of story.

That process does not involve this:

WHINEWHINEWHINEWHINEWHINEWHINEWHINEWHINEWHINEWHIN EWHINEWHINEWHINEWHINEWHINEWHINEWHINEWHINEWHINEWHINE WHINEWHINEWHINEWHINEWHINEWHINEWHINEWHINEWHINEWHINEW HINEWHINEWHINEWHINEWHINEWHINE

So bear that in mind when detecting less-than-perfect English in an online article.


RE: proofreading
By mikeyD95125 on 9/21/09, Rating: -1
RE: proofreading
By lukasbradley on 9/22/09, Rating: 0
RE: proofreading
By adiposity on 9/21/2009 2:52:59 PM , Rating: 2
A typo is one thing, but spelling checks make it worse. The wrong word is substituted and makes it confusing. In this case my knowledge of hdds allowed me to deduce what I assume was the right word ("outer"). However, I did for a second wonder if he meant "every other," which makes no sense of course.

Yes, I do find it annoying, but it's a fact of life. These posts don't have a proof reader, obviously. Welcome to the world of free news.

-Dan


RE: proofreading
By TomZ on 9/21/2009 2:56:39 PM , Rating: 2
DT isn't really free. It is ad-supported, and we "pay" for the site by viewing and clicking on the ads. Furthermore, DT is a for-profit company with employees who I assume are paid for their work. As such, they should be held to professional standards.

Maybe they are all donating their time for free, but I doubt it.


RE: proofreading
By jonup on 9/21/09, Rating: -1
RE: proofreading
By adiposity on 9/21/2009 5:05:24 PM , Rating: 2
I don't have any issue holding them to higher standards. I'm not sure a wrong word is worth getting getting too-riled up about. Mick, on the other hand...

-Dan


"a product that helps people... watch porn."
By Yawgm0th on 9/21/2009 9:55:08 AM , Rating: 5
quote:
The latest version of Seagate's SeaTools software allows for short-stroking,


With the new short-stroking technology, Seagate has further improved its product for porn watchers.




By FITCamaro on 9/21/2009 1:16:03 PM , Rating: 2
Win.

Their next technology on the horizon is known as cupping.


By blwest on 9/21/2009 9:22:03 PM , Rating: 2
However, the spooning operation isn't supported as it takes too long.


By maverick85wd on 9/21/2009 3:41:24 PM , Rating: 2
were it in my power, I'd give you a 6 for that one!


By MrPoletski on 9/24/2009 10:06:11 AM , Rating: 2
Unfortunately, when I short stroke I don't reach the end of my HD.

So that feature isn't really for me...


Do we really need 6 Gpbs?
By jackk on 9/21/2009 11:08:33 AM , Rating: 3
Seagate claims that a sustained transfer rate will be 140MBps (similar to velociraptor speed).
What's the point than of using 6 Gpbs channel?




RE: Do we really need 6 Gpbs?
By Laereom on 9/21/2009 11:12:44 AM , Rating: 5
Burst fire!


RE: Do we really need 6 Gpbs?
By Hieyeck on 9/21/2009 1:26:52 PM , Rating: 3
Do a barrel roll!


RE: Do we really need 6 Gpbs?
By afkrotch on 9/22/2009 2:14:06 AM , Rating: 2
quote:
Seagate claims that a sustained transfer rate will be 140MBps (similar to velociraptor speed).
What's the point than of using 6 Gpbs channel?


new NCQ for QoS for aud/vid streaming
improved power management scheme
one interface for all drives, for lower manufacturing costs

Faster bus speed isn't the only change.


Seagate QA is the worst
By aguilpa1 on 9/21/2009 11:29:50 AM , Rating: 3
Good luck getting all your 2TB of files when the thing quits after 3mo. or less, most likely from a messed up firmware issue the company can not fix.




RE: Seagate QA is the worst
By jarman on 9/21/2009 11:37:57 AM , Rating: 2
Agreed. Seagate has really given Microsoft (360) a run for its money in the hardware failure arena recently.


RE: Seagate QA is the worst
By gstrickler on 9/21/2009 8:36:15 PM , Rating: 2
Firmware != hardware

Firmware is software embedded in non-volatile memory, typically Flash memory these days.


RE: Seagate QA is the worst
By plowak on 9/23/2009 2:59:26 PM , Rating: 2
Duh!


RE: Seagate QA is the worst
By MrPoletski on 9/24/2009 10:07:47 AM , Rating: 2
.... and is one of the few bits of code nowadays that might still be written in a real mans programming language.

Assembly FTW!


RE: Seagate QA is the worst
By afkrotch on 9/22/2009 2:17:09 AM , Rating: 2
Weird, my 1.5 TB Seagate hdds work perfectly. Guess some ppl just don't know how to update firmware.


Short Stroking
By 9nails on 9/21/2009 2:16:54 PM , Rating: 2
As I understand it, short stroking a drive makes the outer 1/3 of the disk available, leaving the remainder of the disk inaccessible. In some areas, this is actually desired. Especially since computer performance can bottle necked on a slow performing hard drive. That means these drives would have about 680 Gigs of high performance disk area if they're short stroked. Not bad at all.




RE: Short Stroking
By Fat32 on 9/21/2009 2:34:09 PM , Rating: 2
quote:
As I understand it, short stroking a drive makes the outer 1/3 of the disk available, leaving the remainder of the disk inaccessible. In some areas, this is actually desired. Especially since computer performance can bottle necked on a slow performing hard drive. That means these drives would have about 680 Gigs of high performance disk area if they're short stroked. Not bad at all.


I'm afraid you're wrong. Short stroking is partitioning drive in the following manner: 1) Primary partition beginning 5-10% of the drive size 2) Extended/secondary partition is the rest 95-90%. As a result your data located on outer 5-10% surface of the disk will be much faster and require shorter movement of the head - thus it is called shortstroking. You still can use the rest of the drive, it won't just disappear.

http://en.wikipedia.org/wiki/Disk_partitioning


RE: Short Stroking
By TomZ on 9/21/2009 2:46:15 PM , Rating: 2
Using the rest of the drive will naturally also degrade the performance relative to leaving it unused, so there is a trade-off there.


RE: Short Stroking
By gstrickler on 9/21/2009 8:32:01 PM , Rating: 2
Actually, short stroking a disk drive is using any method (controller setup, partitioning, etc.) to limit the range of travel of the read write head armature. It's not a specific percentage of the range or storage capacity, nor is is necessarily the outer tracks (although those almost always gives the best performance).

Since the performance benefit is related to the reduction in maximum and average seek time, and because seek time is not linear on modern drives, short stroking will generally need to reduce the maximum head motion by 50% to have a notable performance impact. Since all modern drives record more information on the outer tracks, the "outer half" of the tracks contain about 2/3 of the total capacity of the drive. Therefore, to benefit from short stroking, you generally need to use less than 2/3 of the total capacity of the drive. If you can use less of the capacity, you may see even better performance, although the improvement will diminish as you decrease below using 20% of the total capacity.

A related technique, commonly used on 10k and 15k RPM drives, is to slightly increase the spacing between tracks, which allows reducing the "head settling" time for the "servo tracking" to lock on to a track for an accurate read/write. This must be done on the drive controller, it can't be accomplished using partitioning or other methods like short stroking can. You don't have to do this on a 10k/15k drive, but since those drives are focused on performance, it makes sense to do so. That's a major reason why 10k/15k RPM drives have lower maximum storage capacities.

It should be possible to build a 10k/15k drive with 1.5x-2x the storage capacity, but it would cost some the performance, and performance is the reason to go to a 10k/15k drive in the first place. It might still be a bit faster than a 7200 RPM drive, but it probably wouldn't be a good compromise.


Old tech is old
By dhalilahma on 9/22/2009 5:09:28 AM , Rating: 2
[Done in David Attenborough voice] “ Here we see the common Grammar troll, a highly successful sub-species and indigenous to these parts. Across the sprawling plains of Daily Tech you can see herds of sock puppets roaming and, if you look very carefully, you can see a cave troll peaking from under that bridge”

Anyway, this whole thing is a non-story. Seagate could invent a time machine and come back with SATA 600Gbps, they would still have old tech running at old speeds. At this stage it’s like one of those guys taking a crappy car and sticking a big spoiler on, it might look a little faster but under the hood not much.




No point except hype
By justniz on 9/23/2009 6:05:34 PM , Rating: 2
The bottleneck is still the hard drive itself, not the interface, so there's no point in making the interface faster. It won't improve actual performance any.

They just did this to falsely imply that this drive will perform any faster than drives with a 3Gb/sec interface. (Even 3Gb/sec provides about 100x more bandwidth than any harddrive is capable of using fully).




ok so how about
By MrPoletski on 9/21/09, Rating: -1
RE: ok so how about
By armagedon on 9/21/2009 9:43:04 AM , Rating: 1
Good ! that means that you're buying one of those and will compared them. Let us know what you find ...


RE: ok so how about
By Golgatha on 9/21/2009 9:47:37 AM , Rating: 1
Yeah, or a 300GB 15k Hitachi SAS drive for nearly the same price, compared to a $299 MSRP for this new 2TB drive. I suppose the space would be nice to have for a games install drive or Adobe scratch disc though.


RE: ok so how about
By rninneman on 9/21/2009 9:51:08 AM , Rating: 2
You do realize that the Velociraptor is a mechanical SATA drive, right?


RE: ok so how about
By MrPoletski on 9/22/2009 4:26:59 AM , Rating: 2
quote:
I'll believe that any mechanical SATA disk can keep up with a velociraptor when I see it, not before!


Which part of that statement makes you think otherwise?

ok, maybe I should have squirted the word 'other' in there. I specified mechanical sata to stop poeple going 'eeuuu SSD's will keep up with that, duhhh'


RE: ok so how about
By nvalhalla on 9/21/2009 9:51:05 AM , Rating: 2
Seagate states 140MB/sec, or ~1.1Mbps...


RE: ok so how about
By Redwin on 9/21/2009 10:01:21 AM , Rating: 2
Pretty sure you meant ~ 1.1Gbps


RE: ok so how about
By AnnihilatorX on 9/21/2009 10:10:31 AM , Rating: 2
Actually 140MB/s is 1.4Gb/s in SATA terms because 1byte packet is 10bits in SATA transmission due to error correcting bits.


RE: ok so how about
By nvalhalla on 9/21/2009 1:33:08 PM , Rating: 2
Ahh, you're right. Silly me and my 8bit bytes. And yes, I meant Gbps, not Mbps.


RE: ok so how about
By TomZ on 9/21/2009 1:41:49 PM , Rating: 2
quote:
Actually 140MB/s is 1.4Gb/s in SATA terms because 1byte packet is 10bits in SATA transmission due to error correcting bits.
Close, but not correct.

SATA uses 8b/10b encoding, and what this does is encode all the 8-bit byte values into a 10-bit code. The purpose of this is to balance the number of 0's and 1's on the line and to provide frequent transitions to allow for clock recovery.

There is no error detection or error correction done at that level of the protocol stack.

http://en.wikipedia.org/wiki/8B/10B
http://en.wikipedia.org/wiki/Sata#Encoding


RE: ok so how about
By Reclaimer77 on 9/21/2009 9:58:45 AM , Rating: 1
I don't honestly see the point in making faster and faster standard hard drivers. SSD's are here, they are viable, and they are already cheap enough. In the near future the standard setup will be an SSD for OS/apps, and a larger mechanical hard drive for storage (where peak performance won't matter anyway).

Standard hard drives will, however, be viable for servers for a long time. However any serious server will be running SCSI anyway, so these drives don't apply.


RE: ok so how about
By trisct on 9/21/2009 10:20:54 AM , Rating: 1
SSDs are here, but they are still just around the corner from being a smart buy. There are a number of good technology choices out there, but none that are a reasonable cost ratio. Good SSDs still only come from the higher end of product lines. Once Windows 7 and some file system tweaks get to the mass market, jitter-less controllers are the norm, and the fragmentation problem has been solved, the prices for "good" SSD disks will finally come into a window that is properly affordable. A boot disk that costs much more than $200 is a no-go for most people. The Velociraptors are just at the top of that range, and the only SSDs that you can get for that price either aren't trustworthy or they don't consistently outperform the WD.


RE: ok so how about
By mixpix on 9/21/2009 10:24:23 AM , Rating: 2
SSD storage space is the issue. its still far too expensive to get a drive with amount decent storage space (200GB+).


RE: ok so how about
By Reclaimer77 on 9/21/2009 2:42:48 PM , Rating: 2
Comparing the storage capacities of SSD's to hard drives is idiotic. Nuff said. They aren't for storage. They are for performance.


RE: ok so how about
By afkrotch on 9/22/2009 2:36:17 AM , Rating: 2
That's why they aren't viable. Most users can't even tell what performs better than the other. If it mattered, Raptor/Velociraptors would be sitting in every single desktop.

Most users simply don't care. Until SSDs can match a conventional hard drive in price/storage, it will always stay in the upper teir's for performance nuts. Just like the Raptors/Velociraptors.


RE: ok so how about
By Reclaimer77 on 9/23/2009 12:29:35 PM , Rating: 2
quote:
That's why they aren't viable. Most users can't even tell what performs better than the other. If it mattered, Raptor/Velociraptors would be sitting in every single desktop.


Wrong idiot. Velociraptors, compared to SSD's, have year long seek times. It's not the speed of the driver that matters, it's the near zero seek times that really makes them ideal for OS/app drives. Hard drives can NEVER compete with SSD's. They will always be limited by their moving parts.

Not viable ? Funny, Anandtek seems to think they are plenty viable. You might want to read up on sh#$ before you open your dumb trap.

And please, enough with the "most users" argument. "Most users" today benefit for standard technology that a few years ago idiots like you were saying wasn't for "most users". If we were talking about Extreme Edition CPU's I would agree, but EVERYONE can benefit from SSD's.


RE: ok so how about
By Laereom on 9/21/2009 11:16:31 AM , Rating: 2
Well, a lot of consumers disagree with you on SSDs being vialbe in terms of cost effectiveness. Besides, Seagate doesn't make SSDs -- they have to do whatever they can to leverage their HDD technology to stay relevant.


RE: ok so how about
By TomZ on 9/21/2009 2:22:34 PM , Rating: 2
I'm not sure I agree with you. I think SSDs are something that a lot of people are talking about, but only a small number of people are actually buying. And in the meantime, people are still buying machines and upgrades that include plain-old HDDs.

I'd be curious to know the marketshare of SSDs, but I'd be suprised if it exceeded even just 1% right now.


RE: ok so how about
By Reclaimer77 on 9/21/2009 2:38:19 PM , Rating: 2
Are you ppl dense ? The interface isn't the limiting factor on hard drives, it's the mechanical parts. The new version of Sata will NOT improve standard hard disk performance because the drives will never be fast enough.

It WILL however improve SSD performance.

You people sound like a bunch of VHS junkies who claimed DVD will never get "marketshare".


RE: ok so how about
By TomZ on 9/21/2009 2:54:24 PM , Rating: 2
Two points:

1. 6Gbps will also improve performance of HDDs because it allows faster access to reading from and writing to the DRAM cache. After all, not every I/O request requires a full trip down to the magnetic media. Many requests can be served by the DRAM. After all, that is the purpose of the DRAM in the first place!

2. I specifically didn't say that I don't see SSD market share growing in the future, as you imply. I'm saying I believe the current market share is much less than some techies think it is. The perceived market share is much higher because of all the media/blog attention it gets because of being a new/developing technology.


RE: ok so how about
By TomZ on 9/21/2009 3:12:59 PM , Rating: 2
One more point - you are assuming that the only change in the new interface is the speed, but that is not correct. 6Gbps SATA also adds a number of new features that are also beneficial to HDDs:

• A new Native Command Queuing (NCQ) streaming command to enable isochronous
data transfers for bandwidth-hungry audio and video applications
• An NCQ Management feature that helps optimize performance by enabling host
processing and management of outstanding NCQ commands
• Improved power management capabilities

http://www.sata-io.org/documents/SATA-Revision-3.0...


RE: ok so how about
By MrPoletski on 9/24/2009 9:57:28 AM , Rating: 2
This NCQ will require both hardware and software support no doubt and will also primarily benefit enterprise applications.

Desktop PC's just don't have the Iops load for NCQ to make a significant difference - especially given the non existant seek times of SSD's.


RE: ok so how about
By Silver2k7 on 9/22/2009 2:24:14 AM , Rating: 2
"I don't honestly see the point in making faster and faster standard hard drivers. SSD's are here, they are viable, and they are already cheap enough."

If you want to buy me two 1.6TB bitmicro drives, you can have my old Seagate 1.5TB drives.. its viable yes?


RE: ok so how about
By afkrotch on 9/22/2009 2:32:29 AM , Rating: 2
quote:
I don't honestly see the point in making faster and faster standard hard drivers. SSD's are here, they are viable, and they are already cheap enough.


Viable? $300 for this 2 TB drive or $6000 for four 512 GB SuperTalent SSDs. Yep, sounds viable and cheap enough.

quote:
Standard hard drives will, however, be viable for servers for a long time. However any serious server will be running SCSI anyway, so these drives don't apply.


Except that many servers support SATA or SAS. Many companies will use a backup array using SATA, where performance isn't necessary. They simply require a mass amount of storage and it's on the cheap.


RE: ok so how about
By MrPoletski on 9/24/2009 10:13:30 AM , Rating: 2
quote:
Viable? $300 for this 2 TB drive or $6000 for four 512 GB SuperTalent SSDs. Yep, sounds viable and cheap enough.


Those 4 SSD's in RAID 0 will make your PC so fast you'd think you're running your entire PC from a RAM drive.

That would be nearly 1GB/s transfer rate. You could read your entire windows 7 install in under 5 seconds.

That's officially 'stupid fast'.


6Gb
By Alphafox78 on 9/21/09, Rating: -1
RE: 6Gb
By TomZ on 9/21/2009 10:33:54 AM , Rating: 2
That would be true only if you ignore the RAM cache.


RE: 6Gb
By jonmcc33 on 9/21/2009 11:09:58 AM , Rating: 2
There are quite a bit of NAS storage systems with Celeron processors and it isn't the CPU that is the bottleneck. It's mainly the hard drives themselves.


A somewhat "incorrect" statement
By tungtung on 9/21/09, Rating: -1
RE: A somewhat "incorrect" statement
By gmyx on 9/21/09, Rating: -1
By Laereom on 9/21/2009 11:11:27 AM , Rating: 5
The article doesn't say, 'larger than any other regular HDD'. It says 64MB of cache is 'the largest seen on a regular HDD'. Has more than 64MB of cache been seen on a regular HDD? No. Well, hey, I guess 64MB is the largest.


RE: A somewhat "incorrect" statement
By AnnihilatorX on 9/21/2009 11:54:19 AM , Rating: 3
I was frowning when I read that having same thought as you did but realized afterwards that sentence is factually correct.


RE: A somewhat "incorrect" statement
By TomZ on 9/21/2009 1:52:22 PM , Rating: 1
I agree, and I would take it another step further to say it is a little misleading. It gives the user the false impression that there are not other standard HDDs with that size of cache.

In the world of advertising, that type of language is common. But in less biased writing like news articles, you don't typically see these types of "weasel words" used.


"It's okay. The scenarios aren't that clear. But it's good looking. [Steve Jobs] does good design, and [the iPad] is absolutely a good example of that." -- Bill Gates on the Apple iPad














botimage
Copyright 2014 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki